Related Experiment Videos

Risk prediction models for mortality in patients with cardiovascular disease: The BioBank Japan project

Jun Hata1, Akiko Nagai2, Makoto Hirata3

  • 1Department of Epidemiology and Public Health, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan.

Journal of Epidemiology
|February 1, 2017
PubMed

Insights

New risk prediction models were developed for long-term mortality in patients with chronic cardiovascular disease (CVD). These models utilize traditional risk factors to estimate risks of all-cause and cardiovascular death.

Area of Science:

  • Cardiology
  • Public Health
  • Biostatistics

Background:

  • Cardiovascular disease (CVD) is a primary cause of mortality in Japan.
  • Chronic CVD patients require accurate long-term risk assessment for mortality.

Purpose of the Study:

  • Develop and validate novel risk prediction models.
  • Estimate long-term risks of all-cause and cardiovascular death in chronic CVD patients.

Main Methods:

  • 15,058 chronic ischemic CVD patients (aged ≥40) from BioBank Japan were randomly assigned to derivation (n=10,039) and validation (n=5,019) cohorts.
  • Cox proportional hazards regression was used to develop models based on ten traditional risk factors.
  • Model performance was assessed for 5-year mortality risk in the validation cohort over a median follow-up of 8.55 years.

Main Results:

  • Models for all-cause and cardiovascular death were derived using age, sex, CVD subtype, hypertension, diabetes, total cholesterol, BMI, smoking, drinking, and physical activity.
  • The models showed modest discrimination (c-statistics: 0.703 for all-cause, 0.685 for CVD death) and good calibration in the validation cohort.
  • All-cause and cardiovascular death events occurred in 2962/962 and 1536/481 patients in the derivation/validation cohorts, respectively.

Conclusions:

  • Validated risk prediction models for all-cause and cardiovascular death in chronic ischemic CVD patients have been developed.
  • These models can aid in estimating long-term mortality risk for individuals with chronic CVD.
Abstract

Related Concept Videos